GadflyII
GadflyII
AI & ML interests
None yet
Organizations
Well done nvfp4 quant
3
#1 opened 5 months ago
by
josephbreda
Why Your NVFP4 Model Is Slower Than FP8 on the GB10 (NVIDIA Spark) — And How to Fix It
🤯👍 5
6
#5 opened 5 months ago
by
scottgl
SGLang and MTP
1
#2 opened 4 months ago
by
Michalea
Model requests?
12
#4 opened 5 months ago
by
pathosethoslogos
Fails on a single DGX spark with errors below
1
#2 opened 4 months ago
by
Adrian1234
Update MXFP4 format to compressed-tensors
1
#3 opened 5 months ago
by
mgoin
Here's the vLLM recipe I'm using with 2x RTX Pro 6000
👍 3
17
#1 opened 5 months ago
by
zenmagnets
MMLU PRO Benchmark
3
#3 opened 5 months ago
by
sevapru
vLLM 0.16?
1
#2 opened 5 months ago
by
MMaxHugg
confused response
7
#8 opened 5 months ago
by
jiangyizhi
MTP quality, 47 layer
3
#7 opened 5 months ago
by
Michalea
Upload folder using huggingface_hub
#1 opened 5 months ago
by
GadflyII
Can't deploy by vllm 0.14.1 + transformers
8
#6 opened 6 months ago
by
Butterfly-314
can not run
4
#1 opened 6 months ago
by
aliez-ren
please create mlx version of this
3
#4 opened 6 months ago
by
Narutoouz
Wasn't able to recreate MMLU-Pro benchmarks
5
#5 opened 6 months ago
by
zenmagnets
Request for GLM 4.6V
3
#1 opened 6 months ago
by
SFPLM
GadflyII/GLM-4.7-Flash-NVFP4
15
#3 opened 6 months ago
by
Yu21342
Really appreciate that you ran performance comparison tests with BF16!
3
#2 opened 6 months ago
by
zenmagnets